DBA CPU Benchmark3

➜ Back to Library
XLSExcel example XMLXML import file

Description

Categories
Enginatics
Repository
GitHub
This Benchmark report measures a database server's CPU performance for arithmetic calculations by calculating about 40 million square roots.

example performance for different CPU types:
seconds rows/s CPU
7 285714 AMD Ryzen 9 5950X 16-Core Processor
13 153846
12 166667 AMD EPYC 7742 64-Core Processor
31 64516 Exadata CS X8M
42 47619 Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz
42 47619 Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz
41 48780 Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz
41 48780 Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz
43 46512 Intel(R) Xeon(R) CPU E5-2686 v4 @ 2.30GHz
49 40816 Intel(R) Xeon(R) Platinum 8167M CPU @ 2.00GHz
47 42553 Intel(R) Xeon(R) Platinum 8167M CPU @ 2.00GHz
115 17391 SPARC-T5, chipid 1, clock 3600 MHz

SQL 13 lines

select
avg(sqrt(rownum-0.243)+sqrt(rownum+0.873)) num1,
avg(sqrt(rownum-0.123)+sqrt(rownum+0.537)) num2,
avg(sqrt(rownum-0.211)+sqrt(rownum+0.764)) num3,
avg(sqrt(rownum-0.742)+sqrt(rownum+0.372)) num4,
avg(sqrt(rownum-0.850)+sqrt(rownum+0.683)) num5,
avg(sqrt(rownum-0.563)+sqrt(rownum+0.787)) num6,
avg(sqrt(rownum-0.333)+sqrt(rownum+0.543)) num7,
avg(sqrt(rownum-0.985)+sqrt(rownum+0.481)) num8,
avg(sqrt(rownum-0.122)+sqrt(rownum+0.250)) num9,
avg(sqrt(rownum-0.547)+sqrt(rownum+0.434)) num10
from
dual connect by level<=2000000